can-i-run-this-llm

Can I run this LLM?

Can a AMD Instinct MI300X (192GB) run Qwen3 8B?

Qwen · Apache 2.0 · estimated for a 192.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Qwen3 8B at FP16 · 16.4 GB · 141–212 tok/s on a AMD Instinct MI300X (192GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K3.3 GB269–404 tok/sRuns great
Q3_K_M4.1 GB254–381 tok/sRuns great
Q4_K_M5.0 GB240–360 tok/sRuns great
Q5_K_M5.9 GB229–343 tok/sRuns great
Q6_K6.7 GB217–326 tok/sRuns great
Q8_08.7 GB196–294 tok/sRuns great
FP1616.4 GB141–212 tok/sRuns great

Ran Qwen3 8B on a AMD Instinct MI300X (192GB)? Add your real numbers →